MeltFlow: MeltFlow-VAR

The VAR process uses DC power to strike an arc between the electrode and the ingot surfaces causing the electrode to melt. The molten metal droplets fall into a water-cooled mold. The electrode is continuously advanced as it melts to build up an ingot of improved structure and composition.

MeltFlow-VAR performs a rigorous analysis of the process by considering all the physical phenomena as listed below:

• Use of the two-equation k-e model for an accurate treatment of the turbulent flow in the molten pool
• Convective heat transfer in the molten pool, phase change in the mushy region, and conduction heat transfer in the solidified ingot
• Heat loss from the ingot surface to the mold due to the combined effect of radiation and contact heat transfer
• Effect of ingot shrinkage on the loss of contact heat transfer and current conduction between the ingot and the mold
• Distributions of the electric current, self-induced magnetic field, and Lorentz forces
• Analysis of all phenomena in the growing ingot including initial transients and hot topping
• Effect of magnetic stirring in creating an angular stirring force that produces an angular velocity field and causes additional mixing in the pool due to the centrifugal force
• Macrosegregation of the alloying elements caused by the selective rejection or absorption of the alloying elements by the solid and the redistribution of the elements within the molten metal pool
• Motion and dissolution of inclusions
• Determination of the distributions of the Local Solidification Time (LST), dendrite arm spacings (DAS and SDAS), and interdendritic Rayleigh number that quantifies the probability of freckle formation in superalloys.
• Treatment of temperature-dependent material properties of the alloy

The control-volume method is used for performing the solution of the governing equations in an axisymmetric domain. The computational method incorporates many algorithms that address specific aspects of the VAR process as described below:

• Unified treatment of the molten pool, semisolid, and solidified regions of the ingot
• Transient analysis with a special algorithm to address the growth of the ingot
• Automatic determination of the time step
• Robust treatment of nonlinear heat loss from the top, bottom, and side surfaces of the ingot
• Lagrangian method for the calculation of the trajectories of inclusions

Thus, MeltFlow-VAR provides a robust and efficient calculation of the transient behavior of the ingot during the entire process.

MeltFlow-VAR allows easy creation of a process model by specifying ingot geometry, temperature-dependent alloy properties, and melt schedule through a user-friendly graphical interface. Results of analysis are conveniently examined using Tecplot - a powerful visualization software.

MeltFlow-VAR has been shown to accurately predict the observed pool profiles in superalloy and titanium alloy ingots, and alloy concentrations in titanium alloy ingots in practical VAR processes. It is being actively used in the following manner by leading specialty metals companies to obtain substantial cost-savings in process design:

• Refinement of Melt Schedules
• Investigation of Process Anomalies
• Processing of Titanium alloys, Superalloys, and Steels
• Exploration of Process Variants

The computational method and its application for the analysis of a practical VAR process of a Titanium Alloy is discussed in a technical paper presented at the Ti-2007 Conference. It can be downloaded from the following link:

Computational Modeling of the Vacuum Arc Remelting (VAR) Process Used for the Production of Ingots of Titanium Alloys

Click here for downloading the brochure of MeltFlow-VAR.